Fairness-Oriented Semi-Chaotic Genetic Algorithm-Based Channel Assignment Technique for Nodes Starvation Problem in Wireless Mesh Network

Multi-Radio Multi-Channel Wireless Mesh Networks (WMNs) have emerged as a scalable, reliable, and agile wireless network that supports many types of innovative technologies such as the Internet of Things (IoT) and vehicular networks. Due to the limited number of orthogonal channels, interference between channels adversely affects the fair distribution of bandwidth among mesh clients, causing node starvation in terms of insufficient bandwidth, which impedes the adoption of WMN as an efficient access technology. Therefore, a fair channel assignment is crucial for the mesh clients to utilize the available resources. However, the node starvation problem due to unfair channel distribution has been vastly overlooked during channel assignment by the extant research. Instead, existing channel assignment algorithms either reduce the total network interference or maximize the total network throughput, which neither guarantees a fair distribution of the channels nor eliminates node starvation. To this end, the Fairness-Oriented Semi-Chaotic Genetic Algorithm-Based Channel Assignment Technique (FA-SCGA-CAA) was proposed in this paper for Nodes Starvation Problem in Wireless Mesh Networks. FA-SCGA-CAA optimizes fairness based on multiple-criterion using a modified version of the Genetic Algorithm (GA). The modification includes proposing a semi-chaotic technique for creating the primary chromosome with powerful genes. Such a chromosome was used to create a strong population that directs the search towards the global minima in an effective and efficient way. The outcome is a nonlinear fairness oriented fitness function that aims at maximizing the link fairness while minimizing the link interference. Comparison with related work shows that the proposed FA_SCGA_CAA reduced the potential nodes starvation by 22% and improved network capacity utilization by 23%.

Redundancy Coefficient Gradual Up-weighting-based Mutual Information Feature Selection Technique for Crypto-ransomware Early Detection

Crypto-ransomware is characterized by its irreversible effect even after the detection and removal. As such, the early detection is crucial to protect user data and files of being held to ransom. Several solutions have proposed utilizing the data extracted during the initial phases of the attacks before the encryption takes place. However, the lack of enough data at the early phases of the attack along with high dimensional features space renders the model prone to overfitting which decreases its detection accuracy. To this end, this paper proposed a novel redundancy coefficient gradual up-weighting approach that was incorporated to the calculation of redundancy term of mutual information to improve the feature selection process and enhance the accuracy of the detection model. Several machine learning classifiers were used to evaluate the detection performance of the proposed techniques. The experimental results show that the accuracy of proposed techniques achieved higher detection accuracy. Those results demonstrate the efficacy of the proposed techniques for the early detection tasks.